First Results from the Pierre Auger Project

Transcription

1 First Results from the Pierre Auger Project A new cosmic ray observatory designed for a high statistics study of the the Highest Energy Cosmic Rays. Jim Beatty (Ohio State) for the Pierre Auger Collaboration Colorado, USA (in planning) Mendoza, Argentina (construction underway) Jim Beatty for the Pierre Auger Collaboration 1

2 Outline Description of the Experiment Goals Operation Performance The Model-Independent Energy Spectrum Anisotropy results Photon fraction limit Jim Beatty for the Pierre Auger Collaboration 2

3 The Auger Collaboration Participating Countries Argentina Mexico Australia Netherlands Bolivia * Poland Brazil Slovenia Czech Republic Spain France United Kingdom Germany USA Italy Vietnam * 63 Institutions, 369 Collaborators * Associate Jim Beatty for the Pierre Auger Collaboration 3

4 Science Objectives Cosmic ray spectrum above ev ( = 10 EeV) Measure the shape of the spectrum in the region of the GZK feature Spectrum is predicted to be sharply attenuated at E> ev due to scattering on CMB photons Arrival direction distribution Search for departure from isotropy point sources Composition Light or heavy nuclei, photons, exotics? Design Features High statistics (aperture >7000 km 2 sr above ev in each hemisphere) Actual threshold ev Hybrid configuration surface array with fluorescence detector coverage Full sky coverage with uniform exposure Jim Beatty for the Pierre Auger Collaboration 4

5 The Hybrid Design A large surface detector array combined with air fluorescence detectors results in the unique and powerful design Simultaneous shower measurement allows for transfer of the nearly calorimetric energy calibration from the fluorescence detector to the event gathering power of the surface array. FD duty cycle ~10% SD duty cycle ~100% Different measurement techniques force understanding of systematic uncertainties in each. Reconstruction synergy for precise measurements in hybrid events. A complementary set of mass sensitive shower parameters contributes to the identification of primary composition. Jim Beatty for the Pierre Auger Collaboration 5

6 The Observatory Plan Surface Array 1600 detector stations 1.5 km spacing 3000 km 2 Fluorescence Detectors 4 Telescope enclosures 6 Telescopes per enclosure 24 Telescopes total Jim Beatty for the Pierre Auger Collaboration 6

7 The Surface Array Detector Station Communications antenna GPS antenna Electronics enclosure Solar panels Battery box 3 nine inch photomultiplier tubes Plastic tank with 12 tons of water Jim Beatty for the Pierre Auger Collaboration 7

8 Surface Detector Deployment Jim Beatty for the Pierre Auger Collaboration 8

9 SD tanks are self-calibrating (Atmospheric muons used as a standard candle) Muon Signal (I VEM ) ) Number of evts Single muon peak from 3-fold PMT coincidence 25 ns time bin Time-integrated FADC signal (arbitrary units) Measure all signals in Vertical Equivalent Muon units (VEM). Trigger thresholds expressed in instantaneous current I VEM Dynamic range I VEM Time-integrated signals for energy measurement expressed in Q VEM Jim Beatty for the Pierre Auger Collaboration 9

10 SD triggers Time-over-threshold (TOT) 1-5Hz Long signals Typical shower TOTtrace Threshold 20Hz Fast signals (inclined showers) Central trigger, rate~3000/day look for space-time clusters of triggered tanks Muon pulse from inclined shower Event selection (for current spectrum analysis) look for at least 3 TOT triggers in a compact configuration ~600/day (~0.9/tank/day) Jim Beatty for the Pierre Auger Collaboration 10

11 The Fluorescence Detector 440 pixel camera 3.4 meter diameter segmented mirror Aperture stop and optical filter Jim Beatty for the Pierre Auger Collaboration 11

12 The Fluorescence Detector Los Leones 30 deg x 30 deg x 30 km field of view per eye Jim Beatty for the Pierre Auger Collaboration 12

13 A Stereo Event FD Triggers 100ns time bin Threshold trigger on individual pixels Look for tracks 1 Hz trigger rate per 6 cameras Multi-camera events merged within 2 sec Geometry recon within 5 sec, and passed to central data acquisition system Induces SD readout of tanks within range. Obtain Hybrid events with both fluorescence longitudinal profile and surface detector signals. Jim Beatty for the Pierre Auger Collaboration 13

14 Atmospheric Monitoring and Fluorescence Detector Calibration Atmospheric Monitoring Absolute Calibration Central Laser Facility (laser optically linked to adjacent surface detector tank) Atmospheric monitoring Calibration checks Timing checks Radiosondes for atm profile Drum for uniform illumination of each fluorescence camera Lidar at each fluorescence eye for atmospheric profiling - shooting the shower Jim Beatty for the Pierre Auger Collaboration 14

15 Construction Progress 929 surface detector stations deployed (~780 with electronics and sending triggers) Three fluorescence buildings complete each Jim Beatty for the Pierre Auger Collaboration with 6 telescopes 15

16 Example Event 1 A moderate angle event Zenith angle ~ 48º, Energy ~ 70 EeV Typical flash ADC trace Detector signal (VEM) vs time (ns) Lateral density distribution Flash ADC traces Flash ADC traces Jim Beatty for the Pierre Auger Collaboration 16

17 Example Event 2 A high zenith angle event Zenith angle ~ 60º, Energy ~ 86 EeV Flash ADC Trace for detector late in the shower Lateral density distribution Flash ADC traces Jim Beatty for the Pierre Auger Collaboration 17

18 Example Event 3 A hybrid event Zenith angle ~ 30º, Energy ~ 10 EeV Flash ADC traces Lateral density distribution Jim Beatty for the Pierre Auger Collaboration 18

19 Example Event 3 A hybrid event Zenith angle ~ 30º, Energy ~ 10 EeV Fitted Electromagnetic Shower from Fly's Eye 1985 Jim Beatty for the Pierre Auger Collaboration 19

20 The First Data Set Cumulative number of events Collection period 1 Jan 2004 to 5 June 2005 Zenith angles º Total acceptance 1750km 2 sr yr (~ 1.07 * AGASA) Surface array events (after quality cuts) Current rate - 18,000 / month Total - 150,000 Jim Beatty for the Pierre Auger Collaboration 20

21 The SD Exposure Computation Determine the instantaneous effective surface area by monitoring the detector status in real-time. Multiply the trigger efficiency by the effective surface area of the array. Solid angle-averaged trigger efficiency from simulations Integrate over solid angle and time. Exposure=1750 km 2 sr yr = AGASA*1.07 Jim Beatty for the Pierre Auger Collaboration 21

22 Sky Map of Data set The Auger southern site is at 36 S latitude. A region near the south celestial pole is always in view. Galactic coordinates Coverage of Northern hemisphere is limited. Jim Beatty for the Pierre Auger Collaboration 22

23 Hybrid Events Reconstructed 1800/month Total = 10,000 Mostly at low energies near eyes ~2000 with energies above 1 EeV Jim Beatty for the Pierre Auger Collaboration 23

24 Hybrid Core position and direction resolution is measured with CLF Laser shots Core pos. Angle in SDP SD-Hybrid 2D angle Jim Beatty for the Pierre Auger Collaboration 24

25 Hybrid core position and direction resolution is measured with CLF Laser shots Angular resolution (68% CL): Core pos. Hybrid 0.6 degrees Surface array Angle in SDP 2.2º for 3 station events (E< 4 EeV, θ < 60º ). 1.7º for 4 station events (3<E<10 EeV) <1.4º for 5 or more station events (E>8 EeV) Core position Hybrid 60 meters Surface array: < 200 meters SD-Hybrid 2D angle Jim Beatty for the Pierre Auger Collaboration 25

26 The GZK Feature? AGASA (SD) sees the apparent continuation of the spectrum HiRes (FD) a dropoff in the spectrum above 10^19.5eV. Jim Beatty for the Pierre Auger Collaboration 26

27 FD-alone Energy Measurement Pro: The energy measurement is calorimetric. Energy~ionization loss~tracklength~fluorescence emission Cons: Low duty cycle The aperture is not easily determined. For example, if the atmosphere is dirtier than expected: Energy is underestimated. Exposure (integrated trigger efficiency) is overestimated. With enough time and modest money spent on atmospheric monitoring, this problem can be mitigated. Jim Beatty for the Pierre Auger Collaboration 27

28 SD-alone Energy Measurement Con: The energy measurement technique relies on MC simulations of the expected signal level Assumed hadronic interaction model requires extrapolations of collider data to higher energies and rapidities. Uncertainties are difficult (impossible) to estimate. Pros: high duty cycle Exposure is easily estimated The array trigger efficiency is 100% for large showers! self-calibration with atmospheric muons Jim Beatty for the Pierre Auger Collaboration 28

29 The Auger Model-Independent Approach Use the strengths from each technique: FD(Hybrid) energy, SD statistics, SD exposure. From SD data, reconstruct a stable ground parameter S(1000) (SD signal at 1000m) which is correlated with shower energy Empirically determine the S(1000) Energy conversion Measure the zenith angle dependence of the converter using SD data, and also using Hybrid data Use Hybrid data to: Normalize the converter assuming the FD (hybrid) energy scale Determine the energy dependence of the converter Divide the SD energy histogram by the SD exposure to obtain the measured spectrum. Jim Beatty for the Pierre Auger Collaboration 29

30 The SD fitting function is determined from the data LDF: Distribution of signals versus the core distance r (transverse distance of detector to the shower axis) Use a NKG-like LDF ~ r -β Auger data 1.2< sec θ <1.4 Measured β(e,θ) directly from the data. σ β ~ 10% from previous exps. Jim Beatty for the Pierre Auger Collaboration 30

31 The Ground Parameter S(1000) To determine the shower energy, a single ground parameter S(r) is traditionally chosen to minimize the effects of reconstruction errors, and shower-to-shower fluctuations S( r) is determined from interpolating the data using a LDF with fixed β. However, for most events, precise knowledge of the LDF shape is not important because there exists an optimal core distance R opt at which the different fitted LDFs cross-over. Fits with different LDF slope β r opt= m S(r opt ) is a stable ground parameter. Jim Beatty for the Pierre Auger Collaboration 31

32 A fit to a NKG-like LDF function is used to extract simultaneously S(1000) and the core position Statistical reconstruction error on S(1000). ~1 EeV ~10 EeV Systematic error from LDF shape <4%, estimated by varying LDF slope by +-10%. FADC saturated events have larger r opt Jim Beatty for the Pierre Auger Collaboration 32

33 The total statistical fluctuations of S(1000) including shower-to-shower can be estimated with simulations Protons/QGSJET 10 EeV S(1000) is optimal for a 1.5km spacing array. Jim Beatty for the Pierre Auger Collaboration 33

34 Showers coming from different zenith angles give very different signals due to flux attenuation in the atmosphere θ = 0 θ = g/cm2 = 11 λ I Earth The slant depth is 870g/cm 2 * sec(θ) Jim Beatty for the Pierre Auger Collaboration 34

35 Measure the θ Dependence directly from the data S min dn ds dsin 2 ds θ ToyMC cosθ dω (acceptance correction) Sec(θ) S min The flux dn/ds can in principle be measured independently in each zenith angle bin. Because dn/ds falls monotonically with increasing S, and because the CR flux is isotropic to a good approximation, the contours of constant integrated flux intensity give the θ dependence of the S(1000) Energy Jim Beatty for the Pierre Auger Collaboration 35 relationship.

36 The SD-measured θ dependence (Constant Intensity Cut method) Shape is scanned in θ using bins of sin 2 (θ)=0.1 Normalize at the median zenith angle of 38 degrees. S ( 1000) = S38( E) CIC( θ ) θ Assume for now that CIC(θ) is independent of energy. Note: bins are correlated! 870g / cm CIC( θ ) exp Use intensity cut corresponding to E> ~ ev 2 (secθ sec38 λ atten ) Jim Beatty for the Pierre Auger Collaboration 36

37 Obtain the S38 Energy Correlation with Fluorescence energies from hybrid events log E = log S 38 Strict event selection: tracklength >350g/cm 2 Cherenkov contamination<10% 10EeV Obtain converter: S(1000) / VEM E / EeV = 0.16 CIC( θ ) EeV Note: systematic error grows when extrapolating this rule to 100 EeV! Jim Beatty for the Pierre Auger Collaboration 37

38 Systematic Errors in the FD(Hybrid) Energy Normalization Jim Beatty for the Pierre Auger Collaboration 38

39 Summary of procedure Reconstruct the ground parameter S(1000) Correct for the zenith angle dependence by converting S(1000) to S38 using the measured CIC curve. Convert S38 to Energy using the correlation determined with hybrid data Tank signals S(1000) S38 Energy Each step is empirically determined! Jim Beatty for the Pierre Auger Collaboration 39

40 The Auger Southern Sky Energy Spectrum E/E~30% E/E~50% dn/d(lne) = E*dN/dE Errors on points are Statistical only Systematic errors are estimated at two energy regions Energy measurement (horizontal) Exposure determination (vertical) Jim Beatty for the Pierre Auger Collaboration 40

41 Comparison with HiRes1, AGASA Jim Beatty for the Pierre Auger Collaboration 41

42 Comparison with HiRes1, AGASA-25% Jim Beatty for the Pierre Auger Collaboration 42

43 Our Highest Energy Event E FD ~ ev Landed just outside the array, so not used in spectrum! Jim Beatty for the Pierre Auger Collaboration 43

44 The Top 10 SD events Reconstruction errors only Jim Beatty for the Pierre Auger Collaboration 44

45 Anisotropies It is extremely difficult to measure anisotropies in the sky which populate a narrow energy band in a rapidly falling energy spectrum. The energy search window must be carefully tuned to coincide with the true energy band of any excess. Otherwise the isotropic lower energy background swamps the signal. Systematic errors in the energy measurement can easily contaminate the population of the energy window. Precise angular resolution is needed to detect sources with small intrinsic angular scales. Otherwise the off-source background flux dominates. Furthermore, low statistics require that the energy and angular windows be pre-defined so that the statistical significance of excesses can be evaluated Jim Beatty for the Pierre Auger Collaboration 45

46 Previous Observations of the Galactic Center AGASA 22% excess seen at 4.5σ by AGASA with a partial 20 degree tophat window (centered near the GC) with E=1-2.5EeV. To propagate through the Galactic magnetic fields, the particles are postulated to be neutral, and perhaps to be neutrons from p-p scattering. SUGAR sees a 2.9σ excess with a 5.5 degree window at a slightly different location near the GC with E= EeV. Jim Beatty for the Pierre Auger Collaboration 46

47 Auger: No excess seen in either region Coverage map by shuffling event zenith, day, hour Events smoothed with true resolution, Energy = EeV SUGAR AGASA Significance of excess or deficit Smoothed at SUGAR scale, SUGAR energy window Smoothed at AGASA scale, AGASA energy window Jim Beatty for the Pierre Auger Collaboration 47

48 Search for localized excesses Predefined search parameters: E=1-5 EeV, or E>5 EeV Angular scale=5 degrees, or 15 degrees (tophat) Uses Monte Carlo energy converter instead of CIC (for now) Sky coverage map Look for excesses with tophats centered on each of 50K HEALPIX pixels (1 square degree) Jim Beatty for the Pierre Auger Collaboration 48

49 So far, the data is consistent with isotropy PRELIMINARY Jim Beatty for the Pierre Auger Collaboration 49

50 Other pre-defined targets not seen either with any large significance. Caveats: This analysis still uses the old MC-derived energy converter. Jim Beatty for the Pierre Auger Collaboration 50

51 Photon Limit All top-down models predict a large flux of photon primaries at some energy scale Jim Beatty for the Pierre Auger Collaboration 51

52 For each of 16 selected hybrid events (tracklength>450g/cm2), simulate 100 photon showers Sample the distributions to compute the limit such that in 95% of mock experiments, a better limit is set. Jim Beatty for the Pierre Auger Collaboration 52

53 Photon Fraction Limit Result Photon Fraction <26% at 95%CL for the integrated flux of cosmic rays with E > 10 EeV. Technique is applicable for low statistics datasets at high E. Jim Beatty for the Pierre Auger Collaboration 53

54 Composition Sensitive SD observables Risetime Muon flux Curvature Xmax Jim Beatty for the Pierre Auger Collaboration 54

55 Summary With only 25% of a full Auger-year exposure, we have already: Defined our empirical spectrum analysis strategy and produced our first model-independent spectrum Performed first studies of anisotropies in the sky Defined a procedure for setting photon fraction limits with low statistics. Jim Beatty for the Pierre Auger Collaboration 55

56 Future Plans Complete Auger South by mid 2006 (funding dependent) Full aperture > 7000 km2 sr Fully understand our instruments. Use rapidly expanding data set (x7 in two years) to enable Improved energy assignment Improve LDF measurements Reduce systematic errors in reconstructing events Energy dependent CIC functions Reduce error from extrapolating converter to high energies High statistics study of the trans-gzk spectrum Anisotropy studies and point source searches. Primary composition and hadronic interaction studies Exploit events beyond a zenith angle of 60º search for neutrinos and exotics Begin work on Auger North Jim Beatty for the Pierre Auger Collaboration 56